Kaspa API: $0.03565397 per KAS. Coinbase all-time high: $0.2075.
Proof-of-work mining
Kaspa mining
Price can fall while hash rate climbs. A coin market reprices in minutes; a machine market (ASIC orders, shipping, hosting) answers weeks or months later.
Toccata activated at DAA scoreDifficulty-Adjusted Age: a running block count Kaspa consensus uses instead of wall-clock time, since many blocks land in parallel. 474,165,565, and Rusty Kaspa v2.0.1 is the release. Nothing here is a price call.
Snapshot
What the network is doing
Kaspa API: 338.0 PH/s, about 676xCompared against the 500 TH/s the Kaspa wiki records for November 29, 2022.
Kaspa API: 27.695B KAS mined of a 28.704B KAS maximum.
Kaspa API: 2.18267645 KAS per block, the emission step that began at DAA score 531,207,000.
Dated fallback read from api.kaspa.org on September 11, 2026. These four numbers refresh from the same API when the page loads.
The two markets
Coin market
minutes to days
price, liquidity, sentiment
Machine market
weeks to months
ASIC orders, shipping, hosting, hash rate, difficulty
Reprices KAS within hours.
Price and power cost decide if mining pays.
Profit pulls capital into hardware orders.
Ordered machines arrive after the signal faded.
More machines chase the same reward.
The cold phase sorts who outlasts it.
Cycle model
Coins and ASICs heat and cool at different times
This framing is an outside opinion, not a sourced protocol fact: an X account, @Themooseisloos5, cast the coin-and-ASIC cycle as a four-phase Carnot-engine model.
Four phases
The hot/cold cycle and Kaspa's compressed history
Phase-by-phase breakdown of coin and ASIC market timing.
Hot coins, cold ASICs
Demand moves in before hardware can respond, so margins expand.
Hot coins, hot ASICs
New machines get ordered and hash rate climbs.
Cold coins, hot ASICs
Price cools while ordered machines keep arriving, crushing margins.
Cold coins, cold ASICs
Weak miners shut off, and the system compresses until it finds balance.
Kaspa compressed these phases into one short window instead of Bitcoin's years, which cuts both ways: bigger swings up and down.
Try it
How fast the new supply shrinks
Every block pays a little less, in small steps, not a cliff. Zoom out to see Kaspa's whole issuance history, or zoom in to move around today.
Checking the live network…
today1 year3 years10 years
Each bar is one month's block reward.
This schedule is written into the software every node runs. No vote, company, or founder can add coins to it.
Why the reward falls 5.61% a month with no halving day, and the code that sets it
The reward drops about 5.61% every month, forever, triggered by DAA score rather than the calendar, so there is no single halving day (rusty-kaspa's coinbase.rs). The model uses the live supply when that read succeeds; otherwise it anchors to the dated August 29, 2026 fallback. Other points on the slider are this model scaled to that anchor.
ASICs buy security, with a bill attached
Proof of work turns digital security into a physical bill; an attacker cannot rent cloud compute and be done.
Try it
What an attack on Kaspa would cost
Out-mining everyone else means owning most of the machines. Drag the duration and watch which half of the bill moves.
Reading live hash rates and prices…
1 minute1 day
Compare to another chain, or rent instead of buy
Four proof-of-work chains and two proof-of-stake chains, priced the same way.
Attacking Kaspa for 60 minutes would cost about
reading…
That is the bill for rewriting recent history: undoing payments other people already received.
Limits of this number
Assumptions, exact math, and sources
An attacker needs kHeavyHash machines, electricity, and facilities, every step leaving a trail; that specialization concentrates control among manufacturers and large buyers.
Five signals, the too-fast take, and the slower reading
Five market signals, the reading people jump to, and the slower one
| Signal | Too-fast reading | Slower reading |
|---|---|---|
| Price down | The market lost interest. | Lost demand, miner selling, early-holder profit-taking, broad market weakness, and leverage unwinds can all produce the same line, separately or together. |
| Hash rate down | The network is failing. | Weaker security is one explanation. Inefficient miners leaving after an overheated ASIC cycle is another. Compare against long-term history and attack cost before picking one. |
| Hash rate up while price falls | Security is fine, ignore price. | Delayed ASIC orders can be arriving into a weak market, raising difficulty and squeezing every miner's margin at once. |
| ASIC prices fall | Mining is dead. | Used-machine oversupply after manufacturers and miners overbuilt into a hot phase produces the same price drop. |
| Fees rise | Higher fees are automatically good. | Fees from repeat use people value help miners. Spam and one-off bursts push the number up without meaning the same thing. |
Evidence
Fees are the second revenue source
Issuance pays miners first; fees have to matter later, and whether Kaspa is healthy on that question is not knowable in real time.
Try it
How far fees are from paying for the network
Load a day of real traffic, and compare fee income against what new supply pays out.
Load this block
Kaspa's own traffic. 0.89 tx/s, measured 22 Aug 2026.
Not full yet, so any fee gets you in. Raise transactions per second above to fill it.
Change shape, block rate, price, or timing
Transaction shapeA covenant or a proof carries more mass (Kaspa's measure of block space) than a plain payment, so fewer fit per block.
Plain payment, 1,624 mass each, the lightest shape: up to 307 fit per block.
Block rateBlocks per second is a consensus parameter: a higher rate needs a larger GHOSTDAG k for the same security margin, not just more throughput.
WhenThe block subsidy halves on its own schedule, regardless of traffic.
What happens
307
tx fit per block
3,070
tx/s of room
Every transaction fits, yours included, at 0.89 tx/s against ten blocks a second.
This block, to scale
Each dot is one transaction.
Nothing else is competing for this block.
included
still waiting
your transaction
Fees against the block subsidy
$4.86/day
fees per day
$58,631/day
subsidy per day
At this volume and fee, fees run about 12,054× below the subsidy, on 22 Aug 2026.
New coins pay for nearly all of mining today, and that payment halves every twelve months. Fees are what has to replace it. Real traffic runs about 0.894 tx/s; every other scenario here is hypothetical. Closing the gap takes more throughput, higher fees, or price appreciation.
Show the mass and capacity formulas
Transaction shapes reuse the mass calculator's formula. The mempool draws a weighted random sample by feerate^3, not a strict ranking. Figures are reused exactly from CLAIMS.yml.
What Toccata adds toward fee demand
Toccata widens the design space for fee demand (covenants, ZK proofs, sequencing), but primitives are not usage; the fee side still needs shipped, repeat-use applications.
Checklist
What healthier, worse, and genuinely unknown look like
Signals across hash rate, distribution, tooling, and fees.
Healthier
- Hash rate stabilizes alongside better operator distribution.
- Wallets, explorers, SDKs, and solo-mining paths keep improving.
- Toccata generates repeat app use and fees miners collect.
Worse
- Hash rate keeps falling alongside worse mining concentration.
- Price and volume keep weakening with no demand replacing the hype that left.
- Fees stay irrelevant because apps don't retain users.
Nobody knows live
- How much selling came from miners versus holders versus broader flows.
- Whether future fees can replace enough issuance as the reward declines.
- Whether the next major cycle runs up, down, or sideways.
Solo mining
Point your own ASIC at your own node
An ASIC and a node don't talk directly; the Stratum Bridge translates.
Try it
What a node costs, chain by chain
Node storage
Linear scale: bar length is proportional to gigabytes. A bar under 3% of the row width is floored to stay visible; read its number.
Guess before you look
Kaspa's line is the pruned-node figure measured or estimated from the 1 August 2026 dataset: 50 GB. It is not a universal node requirement; the rising line is a stylized trend, not a forecast.
How the dated 50 GB estimate stays flat: finality, pruning, and the discard window
6 hr
kept
discarded
Finality is where a reorg becomes impossible. Pruning is where a node actually throws the block away. Both depths come from Rusty Kaspa's mainnet parameters.
Finality: 432,000 blocks, 12 hours
Nodes refuse any reorganization that would rewrite the block.
Pruning: 1,080,000 blocks, 30 hours
Where a node discards old data: 648,000 blocks, 18 hours past finality.
How each chain handles history
Archival mode and the unpruned estimate
A node kept forever runs the same software with kaspad's --archival flag. Pruning is verified against a MuHash commitment over the UTXO set (rusty-kaspa's crypto/muhash crate).
ASIC minerDoes the hashing. Its web screen asks for a pool URL, a username, and a password.
Stratum BridgeSpeaks Stratum to the miner and RPC to the node. Neither one talks to the other directly.
Kaspa nodeHands out block templates, and takes solved blocks back.
Kaspa networkA solved block goes out to the DAG, and the reward is paid to the address in it.
Your wallet address travels in the miner's username field, on a computer running a synced node and the Stratum Bridge.
Sources
Sources, with what each backs
Open the table: bridge docs, node guides, and public APIs
| Source | Used for |
|---|---|
| Rusty Kaspa Stratum Bridge docs | Beta status, external-node mode, in-process mode, dashboard, ports, miner setup, and username format. |
| Sample bridge config | The file to download, since the release archives do not carry one. It sets the dashboard port, the Stratum ports, and the node RPC address. |
| Toccata node setup guide | v2.0.1 upgrade guidance, miner and pool checks, fee policy, and post-Toccata field preservation. |
| Rusty Kaspa v2.0.1 release | The Windows, macOS, and Linux archives used in this guide. |
| Kaspa public API: blockDAG state | The mainnet DAA and network read behind the snapshot above. |
| Kaspa public API: coin supply and block reward | The supply and reward figures in the snapshot above. |
| Kaspa public API: network hash rate | The hash-rate reading in the snapshot above. |
| Kaspa wiki timeline | Early hash-rate milestones: 60 MH/s in November 2021, 10 TH/s in June 2022, 50 TH/s in August 2022, and 500 TH/s in November 2022. |
| Kaspa wiki tokenomics | Fair-launch and no-premine context. |
| Coinbase Kaspa price page | All-time-high reference and market-data context. |
| CoinGecko Kaspa page and CoinGecko API | Dated current price read. |
| Kaspa Explained status page | Activation tracked separately from the setup instructions. |
The coin-and-ASIC cycle framing is credited to @Themooseisloos5's X posts. Price, hash-rate, supply, and protocol-status claims use the sources listed above.